# Codebook: Gendered targeting: Do parties tailor their campaign ads to women?


## List of variables for ads_imgs_vids.RData

	•	id - Individual ad identifier
	•	country - Country in which the ad was displayed
	•	parlgov_id - Party identifier from ParlGov database
	•	page_id - Unique identifier for the Facebook page that published the ad
	•	page_name - Name of the Facebook page that published the ad
	•	pg_party_name_short - Party name from ParlGov
	•	pg_party_family_name - Party family classification (e.g., Socialist, Conservative)
	
	•	ad_creation_time - Timestamp of when the ad was created

	•	ad_creative_bodies - Text of the ad (only in .RData file)
	
	•	only_women - Dummy variable indicating if the ad targets only women
	•	fem_audience - Share of female audience
	•	audience_avg_age - Average age of the audience
	
	•	nfaces - Number of faces after excluding background faces (-999 = 4 or more), 
	For videos: Average across all screenshots.

	•	faces_all_fem - Dummy variable indicating that all detected faces are women (0 when no faces), 
	For videos: Only women in all screenshots.

	•	faces_any_fem - Dummy variable indicating that any detected face is a woman (0 when no faces), 
	For videos: Any woman in any screenshot.

	•	faces_all_mal - Dummy variable indicating that all detected faces are men (0 when no faces), 
	For videos: Only men in all screenshots.

	•	faces_any_mal - Dummy variable indicating that any detected face is a man (0 when no faces), 
	For videos: Any man in any screenshot.

	•	faces_age_mean - Mean predicted age of detected faces, 
	For videos: Average across all screenshots.

	•	n_screenshots - Number of screenshots taken from a video ad for analysis
	
	•	is_video - Dummy variable indicating that the ad included a video (as first media)
	•	is_img - Dummy variable indicating that the ad included an image (as first media)


## List of variables for snapchat_ads.RData

	•	id - Unique identifier for each ad.
	•	CreativeUrl - URL to the ad
	•	StartDate - Date when the ad was started
	•	OrganizationName - Name of the organization that created the ad
	•	Gender - Gender of the target audience (empty if not targeted by gender)
	•	AgeBracket - Age range of the target audience

	•	nfaces - Number of faces after excluding background faces (-999 = 4 or more), 
	For videos: Average across all screenshots.

	•	faces_all_fem - Dummy variable indicating that all detected faces are women (0 when no faces), 
	For videos: Only women in all screenshots.

	•	faces_any_fem - Dummy variable indicating that any detected face is a woman (0 when no faces), 
	For videos: Any woman in any screenshot.

	•	faces_all_mal - Dummy variable indicating that all detected faces are men (0 when no faces), 
	For videos: Only men in all screenshots.

	•	faces_any_mal - Dummy variable indicating that any detected face is a man (0 when no faces), 
	For videos: Any man in any screenshot.

	•	n_screenshots - Number of screenshots taken from a video ad for analysis (NA for images)